SentenceTransformer based on CocoRoF/ModernBERT-SimCSE-multitask_v03-retry

Model Details

Model Description

• Model Type: Sentence Transformer
• Base model: CocoRoF/ModernBERT-SimCSE-multitask_v03-retry
• Maximum Sequence Length: 2048 tokens
• Output Dimensionality: 1024 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • misc_sts_pairs_v2_kor_kosimcse

Model Sources

• Documentation: Sentence Transformers Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 2048, 'do_lower_case': False}) with Transformer model: ModernBertModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
  (2): Dense({'in_features': 768, 'out_features': 1024, 'bias': True, 'activation_function': 'torch.nn.modules.activation.Tanh'})
)

Usage

Direct Usage (Sentence Transformers)

Evaluation

Metrics

Semantic Similarity

• Dataset: sts_dev
• Evaluated with EmbeddingSimilarityEvaluator

Training Details

Training Dataset

misc_sts_pairs_v2_kor_kosimcse

• Dataset: misc_sts_pairs_v2_kor_kosimcse at e747415
• Size: 449,904 training samples
• Columns: sentence1, sentence2, and score
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	score
  type	string	string	float
  details	min: 6 tokens mean: 18.3 tokens max: 69 tokens	min: 6 tokens mean: 18.69 tokens max: 66 tokens	min: 0.11 mean: 0.77 max: 1.0

• Samples:

  sentence1	sentence2	score
  주홍글씨는 언제 출판되었습니까?	《주홍글씨》는 몇 년에 출판되었습니까?	0.8638778924942017
  폴란드에서 빨간색과 흰색은 무엇을 의미합니까?	폴란드 국기의 색상은 무엇입니까?	0.6773715019226074
  노르만인들은 방어를 위해 모트와 베일리 성을 어떻게 사용했는가?	11세기에는 어떻게 모트와 베일리 성을 만들었습니까?	0.7460665702819824

• Loss: CosineSimilarityLoss with these parameters:
  json

  1{ 2 "loss_fct": "torch.nn.modules.loss.MSELoss" 3}

Evaluation Dataset

Unnamed Dataset

• Size: 1,500 evaluation samples
• Columns: sentence1, sentence2, and score
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	score
  type	string	string	float
  details	min: 7 tokens mean: 20.38 tokens max: 52 tokens	min: 6 tokens mean: 20.52 tokens max: 54 tokens	min: 0.0 mean: 0.42 max: 1.0

• Samples:

  sentence1	sentence2	score
  안전모를 가진 한 남자가 춤을 추고 있다.	안전모를 쓴 한 남자가 춤을 추고 있다.	1.0
  어린아이가 말을 타고 있다.	아이가 말을 타고 있다.	0.95
  한 남자가 뱀에게 쥐를 먹이고 있다.	남자가 뱀에게 쥐를 먹이고 있다.	1.0

• Loss: CosineSimilarityLoss with these parameters:
  json

  1{ 2 "loss_fct": "torch.nn.modules.loss.MSELoss" 3}

Training Hyperparameters

Non-Default Hyperparameters

• overwrite_output_dir: True
• eval_strategy: steps
• gradient_accumulation_steps: 16
• learning_rate: 8e-05
• num_train_epochs: 10.0
• warmup_ratio: 0.2
• push_to_hub: True
• hub_model_id: CocoRoF/ModernBERT-SimCSE-multitask_v03-distill
• hub_strategy: checkpoint
• batch_sampler: no_duplicates

All Hyperparameters

Training Logs

Framework Versions

• Python: 3.11.10
• Sentence Transformers: 3.4.1
• Transformers: 4.48.3
• PyTorch: 2.5.1+cu124
• Accelerate: 1.3.0
• Datasets: 3.3.0
• Tokenizers: 0.21.0

Citation

BibTeX

Sentence Transformers